Structural connectivity is substantially altered in Alzheimer?s disease (AD). Post-mortem studies show loss of myelinated axons, as well as loss of dendrites which correlates with cognitive severity. Interestingly, the development of AD neuropathology, including amyloid plaque and neurofibrillary tangle (NFT) development appears to occur in brain regions that are characterized by thin myelin. However, the extent to which this information can be leveraged in the preclinical stages of AD to predict future cognitive decline and development of dementia due to AD is unknown. The objective of the proposed renewal project is to determine, in vivo, the extent to which structural disconnection predicts cognitive decline, the temporal and spatial relationship between myelin degeneration and development of AD neuropathology in vivo, and the effect of processes that contribute vulnerability to structural connectivity (i.e. neuroinflammation). The central hypothesis is that that structural disconnection (loss of myelinated axons) is an early and critical feature in the neuropathologic process, is impacted by inflammation, and leads to cognitive decline and dementia due to AD. This will be tested by pursuing three specific aims: Aim 1: Determine the extent to which structural connectivity loss (myelin, axonal, and dendritic degeneration) predicts cognitive decline in preclinical AD. This will be accomplished by utilizing existing and prospectively collected CSF to be assayed for markers of structural connectivity and neural injury. Statistical models will test the extent to which connectivity markers improve prediction of cognitive decline when used with markers of classic AD neuropathology. Aim 2: Determine the temporal ordering of structural connectivity loss in relation to regional plaque/NFT development. This will be accomplished by collecting longitudinal [C11]PIB PET to assess amyloid deposition, [F18]THK5351 PET to assess NFT burden, and quantitative myelin imaging with mcDESPOT MRI to assess myelin degeneration in asymptomatic late-middle-aged adults. Longitudinal modelling on these in vivo data will be used to determine the temporal and spatial ordering of these pathologic processes. Aim 3: Determine the contribution of neuroinflammation to neurodegeneration. Existing and prospectively collected CSF will be assayed for markers of neuroinflammation and neural injury and models will test the extent to which loss of structural connectivity is due to elevated inflammation. We expect the results of this project to provide new knowledge concerning the utility of connectivity markers for predicting cognitive decline in preclinical AD, as well as informing the biology of AD. This in turn is expected to lead to earlier diagnosis of AD, contribute to the development of new prevention and treatment strategies, and reduce the prevalence of AD. This project has a high likelihood of success because it will capitalize on the large amount of existing data collected during the prior project period as well as prospectively collected data in well-characterized preclinical participants in the Wisconsin Registry for Alzheimer?s Prevention study and the Wisconsin Alzheimer?s Disease Research Center.